skip to Main Content

Am trying multi_camera_multi_target_tracking_demo with test video files, running the demo on Ubuntu with:

$ python3.9 multi_camera_multi_target_tracking_demo.py -i ./test_video/test1.mp4 ./test_video/test1.mp4 --m_detector intel/person-detection-retail-0013.xml --m_reid intel/person-reidentification-retail-0277.xml

But I encounter an error:

RuntimeError: Check 'false' failed at src/inference/src/core.cpp:100:
[ NETWORK_NOT_READ ] Unable to read the model: intel/person-detection-retail-0013.xml Please check that model format: xml is supported and the model is correct. Available frontends: paddle pytorch tflite tf ir onnx

From what I understand the script wants onnx format and I am using xml format. Can someone give me a tip on how to redownload onnx format model?

when I cloned the open model zoo repo I used the directions omz_downloader --all and omz_converter --all

2

Answers


  1. Clone the specific branch for Open Model Zoo 2022.3.0 if you’re using OpenVINO 2022.3.0:

    git clone --depth 1 -b 2022.3.0 https://github.com/openvinotoolkit/open_model_zoo.git
    

    Additionally, you can download the individual model using the following command:

    omz_downloader --name person-detection-retail-0013
    omz_downloader --name person-reidentification-retail-0277
    
    Login or Signup to reply.
  2. You seem to have the path of the models slightly wrong, and that’s why the program is not finding them.

    If you call omz_downloader --all from the same location of the python program, then the models will be downloaded to ./intel/MODEL_NAME/FPXX/MODEL_NAME.[xml,bin].

    So, to actually call the program with the correct model paths you should call it like this for example(if using the FP32 version):

    python3.9 multi_camera_multi_target_tracking_demo.py -i ./test_video/test1.mp4 ./test_video/test1.mp4 --m_detector intel/person-detection-retail-0013/FP32/person-detection-retail-0013.xml --m_reid intel/person-reidentification-retail-0277/FP32/person-reidentification-retail-0277.xml
    
    Login or Signup to reply.
Please signup or login to give your own answer.
Back To Top
Search